Edge AI for Jetson and Raspberry Pi

Gemma is Google DeepMind’s family of open-weight AI models for developers building applications that can run on their own hardware. Its range covers compact models for phones and IoT devices alongside larger Gemma 4 models for reasoning on personal computers and servers. Some applications can work offline, keeping model inference on the device. Google AI Studio and Google Cloud are also available for hosted use.

Wyoming Satellite connects a microphone and audio playback device to Home Assistant through the Wyoming protocol. It's for people building a self-hosted voice assistant with a separate device for speaking and listening. The project is archived and no longer maintained; its replacement is Linux Voice Assistant, which uses the ESPHome protocol.

DeepStack is a self-hosted computer vision API for developers adding image analysis to camera systems, home automation or other applications. It runs prebuilt and custom models on your own hardware and works fully offline. Image processing stays on the device or server where you host it, with no cloud service required.

MNN is a lightweight C++ framework for developers who want AI models to run on phones, PCs and embedded devices. It handles inference and training on the device, with a focus on small application footprints and hardware acceleration. The project is open source under Apache 2.0, and Alibaba uses it in apps including Taobao, Youku and DingTalk.

LiteRT is Google's open-source framework for developers building AI into apps that run on users' own devices. It succeeds TensorFlow Lite and covers model conversion, optimization and local inference. It's licensed under Apache 2.0.

ncnn is a C++ framework for developers building on-device AI into mobile, desktop and embedded applications. Its focus is running neural networks with a small memory footprint and no third-party runtime dependencies. Models run on the target device's CPU or a supported Vulkan GPU.

ExecuTorch is PyTorch's runtime for developers building AI into mobile apps, desktop software and embedded devices. It runs models on the user's hardware, with support for Android, iOS, Linux, macOS and Windows, as well as microcontrollers. Developers can reuse a PyTorch model across targets, though hardware-specific deployments need their own exported model files.

MediaPipe is an open-source toolkit for developers adding on-device AI to applications on Android, iOS, the web, desktop and edge devices. It pairs pretrained models with APIs for specific tasks, so developers can use existing solutions or customize them for their applications. The project uses the Apache 2.0 license.

jetson-containers is a Docker container build system for developers running local AI and robotics workloads on NVIDIA Jetson hardware. It supplies prebuilt images and lets you combine AI packages into custom containers, reducing the work of assembling compatible GPU software for JetPack/L4T.

OpenVoiceOS is a free, open-source voice AI platform for people building their own smart speakers or adding voice control to devices. Its core builds on a fork of MycroftAI/mycroft-core, and most classic Mycroft skills also work with it. The project uses the Apache 2.0 license, which permits personal and commercial use.

Cactus is an on-device AI engine for developers building automation into mobile apps, wearables and embedded devices. Its Needle model handles tool calling locally, so a device can turn a request into an action without an internet connection. The focus is small devices, including smart home hardware, robots and microcontrollers.

RKLLM is a software stack for developers building local AI applications on Rockchip hardware. It uses the chip's neural processing unit (NPU) to run language and multimodal models on development boards, with support for the RK3588, RK3576, RK3562 and RV1126B series.

Willow is a self-hosted voice assistant platform for people who want home automation voice control on their own hardware. It runs on Espressif's ESP32-S3-BOX family and connects to Home Assistant, openHAB, or other services that accept speech results over HTTP. The project is open source under Apache 2.0.

Triton Inference Server, offered by NVIDIA as Dynamo-Triton, is a self-hosted AI inference server for teams deploying models in applications. It serves models from different frameworks through one server, with support for on-premises hardware, cloud infrastructure and edge devices. It's open source under the BSD-3-Clause license.

Moondream is a vision model for developers building software that needs to understand images. It can answer questions about a picture, write captions, locate objects, identify points and segment regions. The open-weight models can run on your own hardware, including in an air-gapped environment. The repository code is licensed under Apache 2.0; check each model checkpoint’s own terms for use.

Roboflow Inference is a self-hosted computer vision server for teams building camera and image analysis systems. It runs on your own computer, server, or edge device and combines model predictions with workflows for tracking, counting, measuring, and responding to events. Roboflow also offers hosted servers and a Serverless Cloud API, where processing runs on its infrastructure.

jetson-stats monitors NVIDIA Jetson hardware and gives you control over its power and cooling settings. It runs locally on the board, with a terminal interface called jtop for people developing or running workloads on Jetson devices, including local AI applications.

OpenVINO is an Apache 2.0 licensed toolkit for developers who want to run AI models locally or serve them on their own infrastructure. It converts and optimizes models for inference, with support for x86 and ARM CPUs, Intel integrated and discrete GPUs, and Intel NPUs. Its runtime works on Linux, Windows and macOS.

Ultralytics YOLO is an open-source Python computer vision library for developers building applications that analyze images and video on their own hardware. It supports local and edge deployment, including NVIDIA Jetson, Raspberry Pi and mobile phones. A separate hosted platform provides browser-based annotation, cloud GPU training and managed prediction endpoints.